I've Worked on LLM Inference Optimization at Scale — Who Could Use That Experience?

I've spent several years working on AI infrastructure, LLM inference optimization, and distributed GPU systems at NVIDIA. My experience includes improving inference performance, reducing latency, and supporting reliable AI workloads in production environments. I've worked with Python, Triton Inference Server, TensorRT, CUDA, Kubernetes, and LangGraph. I'm interested in connecting with founders, AI engineers, and technical teams building LLM applications, RAG platforms, or AI agent systems. What I can bring: Practical experience optimizing LLM inference latency and throughput. Guidance on GPU utilization, batching, and model-serving architecture. Experience with distributed AI infrastructure and production reliability. Technical insights into RAG pipelines and multi-step AI agent orchestration. What I'm open to: I'm happy to share technical knowledge, review architecture ideas, explore AI infrastructure challenges, or collaborate on an existing AI project. I'm particularly interested in connecting with teams working on scalable, cost-efficient AI systems. If you're facing challenges with LLM performance or AI infrastructure, I'd be glad to exchange ideas and explore opportunities to work together.